© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 91161
    Prev
    Next
    INDEX
    Explanations

    unlocking potential overthinking

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    able
    0.53
    aceous
    0.48
    Mutable
    0.47
    vdots
    0.46
    磁
    0.45
    shiro
    0.45
    листи
    0.45
     অনুমতি
    0.44
    يل
    0.44
    య
    0.44
    POSITIVE LOGITS
     सीकर
    0.52
    ўцаў
    0.48
     العام
    0.44
     परन्तु
    0.44
     ஏற்பட்டது
    0.42
     هستیم
    0.42
     Blues
    0.41
     streptomycin
    0.41
    ,_-
    0.40
     sought
    0.39
    Activations Density 0.004%

    No Known Activations